# psmarter/CUDA-Practice

<https://repo.masters-masters.com/en/repos/psmarter/CUDA-Practice>

[Home](https://repo.masters-masters.com/en) › [psmarter](https://repo.masters-masters.com/en/owners/psmarter) › [CUDA-Practice](https://repo.masters-masters.com/en/repos/psmarter/CUDA-Practice)

psmarter/CUDA-Practice has 127 stars on GitHub. Its RepoMasters card: Prototype tier. CUDA编程练习项目-Hands-on CUDA kernels and performance optimization, covering…

| Field | Value |
| --- | --- |
| Owner | [psmarter](https://repo.masters-masters.com/en/owners/psmarter) |
| Stars | 127 |
| Forks | 12 |
| Language | [Cuda](https://repo.masters-masters.com/en/languages/cuda) |
| License | MIT |
| Created | 2026 |
| Topics | cuda, cuda-kernels, cutlass, flash-attention, gemm, gpu-programming, high-performance-computing, llm-inference, nccl, nsight-compute |
| Tier | [Prototype (100 to 249 ★)](https://repo.masters-masters.com/en/ranks/rush) |
| View on GitHub | [psmarter/CUDA-Practice](https://github.com/psmarter/CUDA-Practice) |

## Repository

CUDA编程练习项目-Hands-on CUDA kernels and performance optimization, covering GEMM, FlashAttention, Tensor Cores, CUTLASS, quantization, KV cache, NCCL, and profiling.

The tier for repos with 100 to 249 ★. Published odds per card pulled: 70%. Rarity: Common.

## More repos by psmarter

- [psmarter/mini-infer · ★ 237 · Python · Prototype](https://repo.masters-masters.com/en/repos/psmarter/mini-infer)
- [psmarter/PMPP-Learning · ★ 197 · C++ · Prototype](https://repo.masters-masters.com/en/repos/psmarter/PMPP-Learning)

## Nearby in Cuda

- [sjfeng1999/gpu-arch-microbenchmark · ★ 129 · Cuda · Prototype](https://repo.masters-masters.com/en/repos/sjfeng1999/gpu-arch-microbenchmark)
- [Avarok-Cybersecurity/dgx-vllm · ★ 128 · Cuda · Prototype](https://repo.masters-masters.com/en/repos/Avarok-Cybersecurity/dgx-vllm)
- [JustVugg/nanoeuler · ★ 127 · Cuda · Prototype](https://repo.masters-masters.com/en/repos/JustVugg/nanoeuler)
- [dingwentao/GPU-lossless-compression · ★ 126 · Cuda · Prototype](https://repo.masters-masters.com/en/repos/dingwentao/GPU-lossless-compression)
- [lzhengchun/matrix-cuda · ★ 125 · Cuda · Prototype](https://repo.masters-masters.com/en/repos/lzhengchun/matrix-cuda)
- [dgSPARSE/dgSPARSE-Lib · ★ 125 · Cuda · Prototype](https://repo.masters-masters.com/en/repos/dgSPARSE/dgSPARSE-Lib)

---
Data: GitHub · Illustration drawn by RepoMasters from the repo’s ID and its language color.